www.nature.
com/scientificreports
OPEN
Received: 2 October 2017
Accepted: 16 March 2018
Published: xx xx xxxx
SCIeNtIfIC REPOrTS | (2018) 8:5620 | DOI:10.1038/s41598-018-23755-y
Apolipoprotein A-II induces acute-
phase response associated AA
amyloidosis in mice through
conformational changes of plasma
lipoprotein structure
Mu Yang   1,2, Yingye Liu1,3, Jian Dai1, Lin Li1, Xin Ding1, Zhe Xu1, Masayuki Mori1,4,
Hiroki Miyahara1, Jinko Sawashita1,5 & Keiichi Higuchi1,5
During acute-phase response (APR), there is a dramatic increase in serum amyloid A (SAA) in plasma
high density lipoproteins (HDL). Elevated SAA leads to reactive AA amyloidosis in animals and humans.
Herein, we employed apolipoprotein A-II (ApoA-II) deficient (Apoa2−/−) and transgenic (Apoa2Tg) mice
to investigate the potential roles of ApoA-II in lipoprotein particle formation and progression of AA
amyloidosis during APR. AA amyloid deposition was suppressed in Apoa2−/− mice compared with wild
type (WT) mice. During APR, Apoa2−/− mice exhibited significant suppression of serum SAA levels and
hepatic Saa1 and Saa2 mRNA levels. Pathological investigation showed Apoa2−/− mice had less tissue
damage and less inflammatory cell infiltration during APR. Total lipoproteins were markedly decreased
in Apoa2−/− mice, while the ratio of HDL to low density lipoprotein (LDL) was also decreased. Both
WT and Apoa2−/− mice showed increases in LDL and very large HDL during APR. SAA was distributed
more widely in lipoprotein particles ranging from chylomicrons to very small HDL in Apoa2−/− mice. Our
observations uncovered the critical roles of ApoA-II in inflammation, serum lipoprotein stability and AA
amyloidosis morbidity, and prompt consideration of therapies for AA and other amyloidoses, whose
precursor proteins are associated with circulating HDL particles.
Amyloidosis is a group of diseases characterized by extracellular or intracellular deposition of insoluble amyloid
fibrils, which are aggregates formed from normally soluble proteins via conformational changes caused by various
mechanisms1,2. Several serious human diseases such as Alzheimer’s disease, type II diabetes, prion disease and
familial amyloid polyneuropathy (FAP) are associated with amyloid fibril deposition3. Reactive amyloid A (AA)
amyloidosis is a systemic type of amyloidosis and occurs in domestic, laboratory and wild animals, as well as in
humans4–6. AA amyloidosis is a major complication of chronic inflammation in patients with rheumatoid arthritis
and serious infection diseases. As an acute phase plasma protein predominantly synthesized in the liver7,8, serum
amyloid A (SAA) is deposited extracellularly as amyloid fibrils, which leads to tissue structure damage and dys-
function of various organs, including the liver, spleen, kidney and heart, among others9,10.
SAA was first identified as a serum protein that cross-reacts with antibodies against AA protein11–13. During
the acute phase reaction (APR) of inflammation, the concentration of plasma SAA, as a high density lipopro-
tein (HDL) associated apolipoprotein, may increase up to ~1000-fold. SAA is an evolutionarily conserved pro-
tein, but its function has not been completely elucidated. As a biomarker for inflammation, its role in cancer,
1
Department of Aging Biology, Institute of Pathogenesis and Disease Prevention, Shinshu University Graduate
School of Medicine, Matsumoto, 290-8621, Japan. 2Department of Molecular and Cellular Biology, Baylor
College of Medicine, Houston, TX, 77030, USA. 3Institute of Pediatric Research, Children’s Hospital of Hebei
Province, Shijiazhuang, 050031, China. 4Department of Advanced Medicine for Health Promotion, Institute for
Biomedical Sciences, Interdisciplinary Cluster for Cutting Edge Research, Shinshu University, Matsumoto, 290-
8621, Japan. 5Department of Biological Science for Intractable Neurological Disease, Institute for Biomedical
Sciences, Interdisciplinary Cluster for Cutting Edge Research, Shinshu University, Matsumoto, 390-8621, Japan.
Correspondence and requests for materials should be addressed to M.Y. (email: yangmu1021@hotmail.com)
1
www.nature.com/scientificreports/

Figure 1.  The degree of amyloid deposition in whole body (AI) was determined in WT and Apoa2−/− mice at
12 h, 1 d, 3 d and 10 d after co-injection of AgNO3 and AA fibrils. The AI of each mouse is presented (●; WT
and ; Apoa2−/−). AI was significantly lower in Apoa2−/− mice than WT mice at 12 h to 10 d. *, ** significantly
different between WT and Apoa2−/− mice, P < 0.05, P < 0.01 (Mann-Whitney U-test). The table below the
figure represents the number of mice used.

cardiovascular disease, and inflammatory processes remains controversial14,15. However, its adverse role has been
established in the pathogenesis of AA amyloidosis. Sustained high levels of SAA result in tissue deposition of
the N-terminal fragments of SAA as amyloid fibrils. In mice, AA amyloid deposition can be experimentally
induced by multiple injections of silver nitrate (AgNO3), casein or lipopolysaccharide (LPS), resulting in remark-
able elevation and maintenance of high levels of plasma SAA16. Amyloid enhancing factor (AEF) and AA amyloid
fibrils have been used to induce and/or accelerate AA amyloidosis in mice and other animals, such as hens and
rabbits17–19.
In addition to the stimulation of reverse cholesterol transport from extra-hepatic tissue to the liver, HDL is
known for its preventive roles in cardiovascular disease through anti-oxidant and anti-inflammatory effects20,21.
Apolipoprotein A-II (ApoA-II) is the second most abundant protein component of HDL; however, its roles in
HDL function and metabolism remain unclear22. ApoA-II is reported to be more hydrophobic than apolipo-
protein A-I (ApoA-I), and is closely associated with modulation of HDL metabolism and alteration of HDL
conformation by interacting with ApoA-I and other apolipoproteins23–25. In mice, ApoA-II is a serum precursor
of amyloid fibrils (AApoAII) in age-associated systemic amyloidosis (AApoAII amyloidosis)26,27. Our previous
study found that mouse SAA, ApoA-I and ApoA-II interact with each other during AA and AApoAII amyloi-
dosis28,29. It has been reported that during APR, elevated SAA binds to HDL and decreases levels of ApoA-I and
ApoA-II, leading to alteration of HDL particle size11,30,31.
To investigate the potential role of ApoA-II in lipoprotein particle distribution and progression of AA amy-
loidosis, we induced AA amyloidosis by co-injection of AA amyloid fibrils (AEF) and AgNO3 (inflammation
inducer) in wild type (WT), ApoA-II deficient (Apoa2−/−), ApoA-II overexpressing (Apoa2cTg) and ApoA-I defi-
cient (Apoa1−/−) mice. We found that elevation in serum SAA and AA amyloid deposition was significantly
suppressed in Apoa2−/− mice. Moreover, ApoA-II deficiency resulted in dramatic alteration of lipoprotein par-
ticles and redistribution of apolipoprotein in lipoproteins. These results suggest an important role of ApoA-II in
inflammation, lipoprotein metabolism and AA amyloidosis.

Results
AA amyloid deposition was suppressed in Apoa2−/− mice.  After co-injection of AgNO3 and AA
fibrils, tissue sections from various organs were stained with Congo red, and amyloid deposition (amyloid score:
AS and amyloid index: AI) was subsequently determined by green birefringence under polarizing microscopy.
The liver of WT and Apoa2−/− mice and the spleen of WT mice showed AA amyloid deposition 12 h and 1 d after
injection (Fig. 1 and Supplementary Fig. 1). After 3 to 10 d of injection, AA amyloid deposition had expanded
from the liver and spleen to the stomach, intestine, lung and kidney (Fig. 2). ASs in these organs were significantly
less in Apoa2−/− mice than in WT mice (Figs 1 and 2). The degree of amyloid deposition in the whole body (AI)
was significantly reduced in Apoa2−/− mice at 12 h, 1 d, 3 d and 10 d after treatment (Fig. 1).

Elevation of serum and hepatic SAA mRNA expression was suppressed in Apoa2−/− mice.  In
WT and Apoa2−/− mice, serum SAA levels were undetectable at 0 h, but increased dramatically and reached max-
imum levels 2 d after co-injection with AgNO3 and AA fibrils, and then deceased rapidly until being undetectable
at 10 d. In contrast, upregulation of serum SAA was significantly suppressed at 12 h and 1 d after inflammatory
stimulus in Apoa2−/− mice compared with WT mice (Fig. 3). On the other hand, Apoa1−/− mice showed no dif-
ference compared with WT mice (Fig. 3).
To elucidate the mechanism leading to low serum SAA levels in Apoa2−/− mice, real-time PCR was performed
to assess hepatic Saa1/Saa2 mRNA in WT and Apoa2−/− mice (Fig. 4). Prior to experimental manipulation,

SCIeNtIfIC REPOrTS | (2018) 8:5620 | DOI:10.1038/s41598-018-23755-y 2

www.nature.com/scientificreports/

Figure 2.  Amyloid deposition observed in the liver and spleen 10 d after co-injection with AgNO3 and AA
fibrils. Amyloid deposition was detected by green birefringence in Congo red stained tissue (A–D) or brown
positive area with antiserum against AA in immunohistochemically stained tissue (E–H) of WT (A,B,E and F)
and Apoa2−/− (C,D,G and H) mice.

Figure 3.  Serum SAA levels were elevated during induction of AA amyloidosis. WT, Apoa2−/− and Apoa1−/−
mice were co-injected with AgNO3 and AA fibrils and serum were obtained at 0 h, 6 h, 12 h, 1 d, 3 d, 7 d and
10 d after injection from WT mice, Apoa2−/− and Apoa1−/− mice. Serum SAA levels were detected in WT and
Apoa2−/− mice with specific antisera following SDS-PAGE and Western blot (A). The serum concentrations
of SAA were quantitated using a densitometric image analyzer with NIH Image, and serum SAA levels were
represented as ratios to pooled standard serum isolated from C57BL/6 J mice 1 d after co-injection with AgNO3
and AA fibrils (B). Each symbol represents mean ± SEM. **P < 0.01, ***P < 0.001, WT vs Apoa2−/−, #P < 0.05,
###
P < 0.001, Apoa1−/− vs Apoa2−/− (Turkey–Kramer method of multiple comparisons at corresponding time).
The table below the figure represents the number of mice used.

Apoa2−/− mice expressed a lower Saa1/Saa2 mRNA level compared with WT (P < 0.05; Fig. 4). Furthermore,
Apoa2−/− mice also showed significantly lower SAA levels under inflammatory stimuli at 1 d (P < 0.05).

Pathological damage in lungs was suppressed in Apoa2−/− mice.  Lungs of Apoa2−/− and WT
mice at 12 h to 10 d after treatment with AgNO3 and AA fibrils were evaluated microscopically. Apoa2−/− mice

SCIeNtIfIC REPOrTS | (2018) 8:5620 | DOI:10.1038/s41598-018-23755-y 3

www.nature.com/scientificreports/

Figure 4.  Suppressed hepatic Saa1/Saa2 mRNA expression levels in Apoa2−/− mice during the acute
inflammation stage. mRNA expression levels of Saa1/Saa2 in the liver of WT and Apoa2−/− mice during the pre
(0 h) and acute inflammation (12 h and 1 d) stages were determined by real time-PCR analysis. Expression levels
are represented as ratios to a WT mouse at 0 h. Hepatic mRNA expression levels of SAA were suppressed in
Apoa2−/− mice compared with WT mice in the pre-inflammation stage. During acute inflammation, expression
rates increased dramatically, by as much as ~350 times in WT mice, but were suppressed (~170 times) in
Apoa2−/− mice at 1 d after injection. Each column and bar represents mean ± SEM. *P < 0.05, WT vs Apoa2−/−
(unpaired Student’s t test with Bonferroni Correction for multiple comparisons).

Figure 5.  Pathological damage was examined in the lungs of WT (A,C,E,G and I) and Apoa2−/− (B,D,F,H
and J) mice during inflammation caused by co-injection with AgNO3 and AA fibrils. The level of pathological
damage in the lungs of Apoa2−/− mice was less severe than that of WT mice (HE × 100). Each scale bar indicates
100 μm. The damage score of each mouse is presented (K). There are significant differences in damage score
between WT and Apoa2−/− mice. *P < 0.05, significantly different between WT and Apoa2−/− mice (Mann-
Whitney U-test). The table on the right side of the figure represents the number of mice used.

experienced less tissue damage and less inflammatory cell infiltrates compared to WT mice at 12 h during APR
(Fig. 5C and D). At 1 d during APR, Apoa2−/− mice had less emphysematous changes than WT mice (Fig. 5E and F).
At 3 d during APR, the lungs of WT still exhibited infiltration of inflammatory cells, while Apoa2−/− mice demon-
strated recovery from damage (Fig. 5G and H). At 10 d after APR, both WT and Apoa2−/− mice showed recovery
from inflammatory damage (Fig. 5I and J). Significant differences in damage scores were noted between WT and
Apoa2−/− mice at 12 h to 3 d (Fig. 5K).

SCIeNtIfIC REPOrTS | (2018) 8:5620 | DOI:10.1038/s41598-018-23755-y 4

www.nature.com/scientificreports/

Figure 6.  AA amyloid deposition and serum SAA levels were accelerated in Apoa2Tg mice. The positive
areas of amyloid deposits in the liver and spleen were determined as IHC positive area with antiserum against
AA at 3 and 10 d after co-injection with AgNO3 and AA fibrils (A and B). Three areas in each of the liver and
spleen sections were randomly captured under x200 magnification, and the positive areas (ratio to whole liver
and spleen) were calculated with the ImageJ software. Each column represents the mean at 12 h, 1 d, 3 d, and
each column and bar represents the mean ± SEM. The tables below the figures represent the number of mice
used. *P < 0.05, WT vs. Apoa2Tg (unpaired Student’s t-test). WT and Apoa2Tg mice were co-injected with
AgNO3 and AA fibrils. Serum was obtained at 0 h, 12 h, 1 d, 2 d, 3 d, 5 d, 7 d and 10 d after injection, and serum
SAA levels were determined by Western blot analysis (C and D). Serum SAA levels are represented as a ratio
to pooled standard serum isolated from C57BL/6 J mice. Each symbol represents mean ± SEM. *P < 0.05,
**P < 0.01, WT vs Apoa2Tg (unpaired Student’s t test). The table below the figure represents the number of
mice used.

AA amyloid deposition and elevated serum SAA levels were accelerated in Apoa2Tg mice.  We
compared AA amyloid deposition and serum SAA levels in WT (R1.P1-Apoa2c) mice with those in Apoa2Tg
mice. The serum concentration of ApoA-II in ApoA2Tg was 1.26 times that in WT mice32. In Apoa2Tg mice,
AA immunohistochemistry (IHC) positive area was more abundant in the liver and spleen at 3 d and 10 d
after co-injection compared with WT mice. Significant differences at 10 d were noted in the liver and spleen
(Fig. 6A and B). The amyloid deposition stained positively in IHC with anti-AA antiserum, but negatively with
anti-AApoAII antiserum (data not shown). Apoa2Tg mice also showed significantly higher serum SAA levels at
12 h and 1 d after co-injection compared with WT mice (Fig. 6C and D).
These results suggest that increased levels of serum ApoA-II may accelerate the APR associated elevation of
serum SAA levels and accelerate AA deposition.

Lipoprotein particles exhibited smaller sizes in Apoa2−/− mice.  To elucidate the effect of ApoA-II
deficiency on lipoprotein particles and the distribution of major apolipoproteins containing SAA, we analyzed
HDL particle size by non-denaturing gradient PAGE with serum pre-stained by Sudan Black B (Fig. 7A). Further,
we performed Western blot analysis of serum amyloidogenic apolipoproteins (SAA, ApoA-I, ApoA-II and ApoE)
(Fig. 7B–E). We marked the size classes of HDL1, HDL2, and HDL3 based on the distributions of ApoA-I in
WT mice according to our previously reported results29,32. The predominant form of HDL in WT mice was
HDL3. The HDL particle size increased within 24 h of APR, before returning to the pre-inflammatory size by
72 h. In pre-inflammatory Apoa2−/− mice, the HDL band was weak, broader and smaller corresponding to HDL3
(Fig. 7A), while during APR, the density, size and distribution of HDL particles increased. Without inflamma-
tory stimulation, SAA was not detected by Western blot in WT and Apoa2−/− mice (Fig. 7B). During APR, SAA
increased dramatically and was mainly present within HDL3, HDL2 and HDL1, with the HDLs smaller than
HDL3 in WT mice. However, SAA was found in HDL2 and HDL1 in Apoa2−/− mice, as well as in HDLs smaller
than HDL3 and larger lipoproteins. ApoA-I was mainly found within HDL3 and HDL2 for WT mice, and par-
tially in HDL1, as determined by PAGE. During APR, the level of ApoA-I decreased gradually and was largely

SCIeNtIfIC REPOrTS | (2018) 8:5620 | DOI:10.1038/s41598-018-23755-y 5

www.nature.com/scientificreports/

Figure 7.  ApoA-II deficiency resulted in the disruption of HDL structure and redistribution of SAA during
APR. HDL particle size and distributions of major apolipoproteins among HDL species were analyzed by non-
denaturing 5–15% PAGE. To detect HDL particle size, 5 μl of pooled serum obtained from WT or Apoa2−/−
mice during APR at 0 h, 12 h, 1 d and 3 d after co-injection with AgNO3 and AA fibrils were pre-stained with
Sudan Black B and applied to non-denaturing 5–15% PAGE (A). SAA (B), ApoA-I (C), ApoA-II (D), and ApoE
(E) protein distributions among lipoprotein species were detected by Western blot analysis with specific antisera
following non-denaturing 5–15% PAGE of 1 μl pooled serum of mice. All blots were obtained under the same
experimental conditions, and cropped images of the blots are shown.

found in HDL3 and HDL2, and the particle size only increased slightly. In Apoa2−/− mice, the amount of ApoA-I
was dramatically decreased, and was mainly found in HDL particles smaller than HDL3 in normal and APR
states (Fig. 7C). ApoA-II was distributed mainly in smaller HDL3 compared with ApoA-I containing HDL3 in
WT mice. The amount of ApoA-II decreased and the size of HDL containing ApoA-II increased slightly during
APR (Fig. 7D). There was no ApoA-II in Apoa2−/− mice, as expected. ApoE was distributed widely, and found
in lipoproteins ranging from HDL1 to larger lipoprotein particles in both WT mice and Apoa2−/− mice. These
observations suggest that ApoA-II deficiency resulted in the disruption of HDL structure and redistribution of
elevated SAA and ApoA-I during APR. There was no obvious influence on the distribution of ApoE in this study.
To further confirm these results, pooled sera from WT and Apoa2−/− mice isolated at 0 h, 12 h, and 1 d after
co-injection were analyzed with a HPLC gel filtration system (Fig. 8A). Under normal conditions, HDL choles-
terol levels were markedly decreased in Apoa2−/− mice, while low density lipoprotein (LDL) cholesterol levels
were increased compared with WT mice. During APR at 1 d, WT and Apoa2−/− mice showed an increase in lipo-
protein cholesterol levels in LDL and very large HDL. Western blot analysis of isolated HPLC fractions showed
that SAA protein was clearly associated with very small to very large HDL in WT mice, as well as very small LDL.
However, in Apoa2−/− mice, SAA distributed more widely, and was found in very small to very large HDL, very
small LDL, very low density lipoprotein (VLDL) and chylomicron (CM), compared with WT mice (Fig. 8B).

Discussion
In this study, we investigated whether apoA-II could influence the metabolism of SAA and HDL and formation
of AA amyloidosis in mice. Our previous study showed that serum ApoA-II and SAA interact with AApoAII
and AA amyloid fibrils, and facilitate amyloid formation in R1.P1-Apoa2c mice28. SAA protein cross-reacts with
pre-existing AApoAII amyloid fibrils and complements the seeding effect of AA fibrils to SAA, increasing AA
amyloidosis28. On the other hand, increased SAA expression during APR reduces the serum concentration of
ApoA-II and suppresses AApoAII amyloidosis. However, there are no studies demonstrating the effects of defi-
ciency and overexpression of ApoA-II on SAA metabolism and AA amyloidosis.
We employed Apoa2−/− and Apoa2Tg mice to explore the mechanism by which ApoA-II influences SAA
metabolism and AA amyloidosis. After co-injection of AgNO3 (inflammation inducer) and AA fibrils (seed),
we showed a significant effect of ApoA-II on elevation of serum SAA levels and AA amyloidosis (Figs 1, 2, 3 and
5). To further elucidate the mechanism of SAA suppression in Apoa2−/− mice, we analyzed hepatic SAA mRNA
expression by real time PCR. Expression of hepatic SAA mRNA in Apoa2−/− mice was significantly lower than in
WT mice in both pre-inflammatory and inflammatory states (Fig. 4). During inflammation, hepatic synthesis of
SAA is induced by the macrophage-secreted cytokines IL-6, IL-1, and TNFα via an Nf-κB-dependent pathway.
Pathological investigation showed that Apoa2−/− mice experienced less lung tissue damage and inflammatory
cell infiltration during APR. Moreover, the lungs of Apoa2−/− mice showed quicker recovery from inflammatory
damage than WT mice (Fig. 5A to K). However, the impact of elevated SAA on inflammation and tissue damage
remains controversial8. It has been shown that ApoA-II plays multiple and controversial roles in influencing
inflammation23,34. Some studies have shown that overexpression of ApoA-II could confer pro-inflammatory prop-
erties to HDL by reducing protection against LDL oxidation and stimulating lipid hyperoxidation and monocyte
transmigration in mice34 and that ApoA-II may maintain host responses to lipopolysaccharide (LPS) by suppress-
ing inhibitory activity of LPS binding protein35. In contrast, other studies have suggested an anti-inflammatory
effect of ApoA-II in humans36. Macrophages and monocytes activated by inflammation may cleave the C-terminal
part of SAA and induce AA amyloid fibril formation37. Macrophages may also resorb amyloid deposits37–39. Thus,
inflammation may change the microenvironment of tissues in which amyloid deposits form, and may modulate

SCIeNtIfIC REPOrTS | (2018) 8:5620 | DOI:10.1038/s41598-018-23755-y 6

www.nature.com/scientificreports/

Figure 8.  SAA was more widely distributed in various lipoprotein particles in Apoa2−/− mice as determined
by HPLC analysis. Pooled serum (5 µl) from WT and Apoa2−/− mice at 0 h, 12 h and 1 d after co-injection with
AgNO3 and AA fibrils was analyzed by HPLC for serum lipoproteins (A). Isolated HPLC fractions of ~30 μl
pooled serum from WT and Apoa2−/− mice during APR at 12 h was separated by SDS-PAGE, and SAA protein
was determined by Western blot analysis (B). All blots were obtained under the same experimental conditions,
and cropped images of the blots are shown.

the progression of AA amyloidosis. Our study demonstrates that ApoA-II provokes inflammation and increases
cytokines and macrophages. Further research is required to validate these results.
In mouse AA amyloidosis, amyloid deposition occurs independent of inflammation, while the time and
degree of amyloid deposition is determined by plasma SAA concentration in transgenic mice40,41. In human
AA amyloidosis, plasma SAA concentration is a major factor determining amyloid deposition37. In other sys-
temic amyloidosis, circulating concentrations of precursor protein determine amyloid deposition. In Apoa2Tg
mice, an increase in serum ApoA-II levels of 1.26× was shown to lead to accelerated AApoAII amyloid deposi-
tion32, and reduced ApoA-II concentration was associated with decelerated amyloid deposition by treatment with
calorie-restriction42. Thus, we believe that a lower SAA concentration suppresses AA amyloidosis in Apoa2−/−
mice, and increased SAA accelerates amyloid deposition in Apoa2Tg mice. The genetic background of Apoa2Tg
mice is the R1.P1-Apoa2c strain, and elevation of SAA concentration during APR in this strain was less obvious
compared with C57BL/6 J mice that have a genetic background of Apoa2−/− (Figs 3 and 6). Due to the different
genetic background and the relatively smaller number of Apoa2Tg mice examined compared with Apoa2−/− mice,
we plan to examine transgenic ApoA-II mice more intensively in future studies to provide additional support for
our findings. We believe that a decrease in serum SAA levels in Apoa2−/− mice was associated with suppression
of amyloid deposition and tissue damage. However, there is a possibility that structural changes in AA amyloid
fibrils of Apoa2−/− mice could affect amyloid deposition, and will be the topic of future amyloidosis research.
During APR, plasma proteins increase by at least 25% and the major acute-phase proteins include C-reactive
protein (CRP), SAA and fibrinogen43. Increased SAA circulates with HDL and results in redistribution of HDL
and its apolipoproteins. HDL contains several apolipoproteins, such as SAA, ApoA-I, ApoA-II, ApoA-IV, C-II,
C-III and ApoE, which are currently considered to be amyloidogenic or amyloid-associated proteins3,44. The
most abundant HDL apolipoprotein that is impacted during APR is ApoA-I45–47. A number of reports have found
marked decreases in serum ApoA-I during APR48–51, and acute phase HDL is larger in size than normal HDL3,
with the radius extending into the HDL2 range52. In contrast, other studies reported no changes in HDL cho-
lesterol or ApoA-I levels when SAA was increased to levels comparable to those during infection or inflamma-
tion46,53. In our study, HDL particle size was increased and ApoA-I and ApoA-II decreased during APR (Fig. 7).
Some reports have revealed decreased hepatic expression and serum concentrations of ApoA-II during the acute
phase28,43. Interestingly, our observation showed that the increased SAA was widely distributed during APR, from
LDL, HDL1, HDL3 to very small HDL, while ApoA-I and ApoA-II were not found in LDL and very small HDL
(Figs 7 and 8).
Despite these results, the role of ApoA-II in HDL metabolism remains unclear22. We sought to elucidate the
effect of ApoA-II deficiency on the distribution of lipoprotein particles and other apolipoproteins during APR.
In Apoa2−/− mice, lipoproteins were dramatically decreased and widely distributed, from LDL to very small HDL
(Figs 7 and 8). ApoA-I was also dramatically deceased, with its distribution ranging from HDL2 to very small

SCIeNtIfIC REPOrTS | (2018) 8:5620 | DOI:10.1038/s41598-018-23755-y 7

www.nature.com/scientificreports/

HDL. HDL cholesterol in Apoa2−/− mice exhibited a smaller size compared with that of WT mice (Fig. 8A).
ApoA-II-deficient mice have been reported to have dramatically decreased and smaller HDL54, consistent with
our observations. In contrast, Apoa1−/− mice showed slight decreases in ApoA-II levels and a larger HDL size in
our previous study29. During APR, Apoa2−/− mice showed an increase in lipoprotein levels, mainly from LDL
and very large HDL at 24 h (Fig. 8A). Surprisingly, we observed that SAA in Apoa2−/− mice was more predom-
inantly located in CM, LDL, HDL and very small HDL, as assessed by PAGE and HPLC gel filtration analysis
(Fig. 8B). However, ApoA-I was found mainly in very small HDL during APR in Apoa2−/− mice. Our previous
study showed that ApoA-I plays a key role in maintaining ApoA-II distribution and HDL particle size29. On
the other hand, it has been reported that ApoA-II is an important factor regulating HDL size and the ratio of
ApoA-I to ApoA-II in plasma55. ApoA-II is more hydrophobic and modulates more strongly the conformation
of HDL than ApoA-I56–58. Our results revealed that, compared with ApoA-I, ApoA-II is as a stronger regulator of
lipoprotein metabolism and apolipoprotein stabilization. Conformational changes of lipoprotein particles may
alter the binding strength of SAA to lipoprotein. The extracellular matrix bound by SAA includes heparan sulfate
proteoglycan and anti-inflammatory effects of lipoproteins. It has been suggested that conformational changes of
lipoproteins may increase the susceptibility to AA amyloidosis59.
Our study showed that A-II plays a critical role in the pathogenesis of AA amyloidosis in mice. Important
factors associated with the pathogenesis of AA amyloidosis include: (1) serum concentration of SAA, (2) SAA
associated circulating lipoprotein structure, and (3) the microenvironment in which SAA forms fibrils and depos-
its. Recent studies showed that ApoA-II plays multiple roles in regulating the metabolism of plasma HDL and its
apolipoproteins, and in prevention or acceleration of cardiovascular disease23,54,60. This study sheds light on the
relationship between ApoA-II and SAA, and provides new information regarding the pathogenesis of amyloidosis
associated with HDL-related proteins. ApoA-II may serve as a therapeutic target for AA. Additional studies are
warranted to further explore this important issue.

Materials and Methods

Mice and induction of AA amyloidosis.  C57BL/6 JJmsSlc mice were obtained from Japan SLC Inc.
(Hamamatsu, Japan). C57BL/6-B6.129P2-Apoa1tm1Unc/J and C57BL/6-B6.129S4-Apoa2tm1Bres/J mice were pur-
chased from Jackson Laboratories (Bar Harbor, ME). The ApoA-I deficient (Apoa1−/−) and ApoA-II deficient
(Apoa2−/−) strains were produced by backcrossing the Apoa1tm1Unc and Apoa2tm1Bres allele 10 times to C57BL/6J
mice. Apoa2c transgenic mice (Apoa2Tg) were generated on a genetic background of congenic SAMR1.SAMP1-
Apoa2c (R1.P1- Apoa2c) mice using backcross procedures32. Mice were maintained under specific pathogen
free (SPF) conditions at 24 ± 2 °C with a light-controlled regimen (12 hours light/dark cycle) in the Division of
Laboratory Animal Research, Research Center for Supports to Advanced Science, Shinshu University. A commer-
cial diet (MF; Oriental Yeast, Tokyo, Japan) and tap water were provided ad libitum. All experiments using ani-
mals were performed with the approval of the Committee for Animal Experiments of Shinshu University under
permit numbers 270016 (from 2015) and 280014 (from 2016), and with the approval of the Shinshu University
Safety Committee for Recombinant DNA Experiments under permit numbers 15-007 (from 2015) and 16–016
(from 2016). Approved protocols were strictly followed.
Two-month-old male mice were subjected to the experiments for APR and AA amyloidosis. Mice were
co-injected with AgNO3 (1%, 0.5 ml, subcutaneous injection) and AA amyloid fibrils (100 μg, intravenous injec-
tion). Blood samples were collected at 6 and 12 hours (h), and at 1, 2, 3, 5, 7 and 10 days (d) after injection. Mice
were sacrificed by cardiac puncture under isoflurane anesthesia 12 h, 1 d, 3 d, and 10 d after co-injection, and
major organs were fixed in 10% neutral buffered formalin and embedded in paraffin.

Isolation of amyloid fibrils.  AA fibrils were isolated from the pooled livers and spleens of C57BL/6 J mice
with severe AA amyloidosis. Amyloid fibril fractions were isolated by Pras’ method with some modification32.
Isolated amyloid fibrils were suspended in deionized/distilled water (DW) at a concentration of 1.0 mg/ml and
were stored at −70 °C until use. We placed 1 ml of this mixture into a 1.5 ml Eppendorf tube and sonicated on ice
for 30 seconds (s) with an ultrasonic homogenizer VP-5S (Tietech Co., Ltd., Tokyo, Japan) at maximum power.
This procedure was repeated 5 times at 30 s intervals. Sonicated AA fibril samples were used immediately.

Detection of amyloid deposition and inflammation in mice.  Mouse organs embedded in paraffin were
sliced into 4 μm sections and stained with Mayer’s hematoxylin/eosin (HE) or Congo red. Amyloid deposition was
identified under polarizing microscopy by green birefringence in tissue sections stained by Congo red61. AA fibrils
were also identified by IHC analysis using the avidin–biotin horseradish peroxidase complex method with specific
rabbit antiserum against mouse AA, which was produced against guanidine hydrochloride-denatured AA fibrils in
our laboratory62. The antiserum reacts specifically with AA amyloid deposits in IHC analysis and serum SAA protein
in Western blot analysis32,63,64. For quantitative IHC analysis, 3 areas in each liver and spleen section were randomly
captured and the ratios of positively stained areas with anti-AA antiserum to the whole section areas were calculated
using an image processing program (NIH Image J software, version 1.61). The intensity of the AA deposit was also
determined semi-quantitatively using the amyloid score (AS) and amyloid index (AI). As described previously, the
AI is the average of the AS graded from 0 to 4 in the 7 organs examined (liver, spleen, skin, heart, stomach, small
intestine and tongue) in sections stained with Congo red and observed under polarizing microscopy63. Two blinded
observers, who had no information regarding the tissues, graded the AS.
Two blinded observers, who had no information regarding the tissue, observed pathological damage and inflam-
matory cell infiltration in tissue specimens stained with HE. We evaluated the damage score by quantifying the area
of inflammatory cells and pulmonary edema. The inflammatory cell infiltration area was scored as follows: 0 (<5%),
1 (5–25%), 2 (25–50%), 3 (>50%). The area of pulmonary edema was scored as: 0 (absent), 1 (<25%), 2 (25–50%), 3
(>50%). The sums of these two area scores represented final damage scores for statistical analyses.

SCIeNtIfIC REPOrTS | (2018) 8:5620 | DOI:10.1038/s41598-018-23755-y 8

www.nature.com/scientificreports/

Serum concentration of SAA.  We isolated serum by centrifugation of blood at 3,000 g for 20 min at
4 °C. The serum (0.5 or 1 μl) was boiled in sample buffer (2% SDS, 12% glycerol, 62.5 mM Tris pH 6.8, 10%
2-mercaptoethanol, 0.02% bromphenol blue) and separated by electrophoresis at 2 mA for 2 h, followed by 20 mA
for 4 h on Tris-Tricine/SDS-16.5% polyacrylamide gel electrophoresis (SDS-PAGE). Proteins in the gels were
transferred electrophoretically to polyvinylidine difluoride (PVDF) membranes. Proteins on the membranes
were reacted with anti-AA antiserum (1:3000), followed by peroxidase-conjugated anti-rabbit IgG solution
(1:3000) (Cell Signaling Technology, Massachusetts, USA)63. Immunoreactive proteins were visualized with the
enhanced chemiluminescence (ECL) system (Amersham Biosciences, Buckinghamshire, England) with X-ray
film (Amersham Biosciences). For quantification, Western blot images were captured and analyzed using Scion
Image version 4.0.3.2. Serum SAA levels were represented as ratios to pooled standard serum (ratio = 1.0) isolated
from C57BL/6 J mice at 1 d after co-injection of AgNO3 and AA amyloid fibrils.

Saa1 and Saa2 mRNA expression in the liver.  Total RNA was extracted from the livers of C57BL/6 J and
Apoa2−/− mice at 0 h, 12 h and 1 d after co-injection of AgNO3 and AA amyloid fibrils, using TRIZOL Reagent
(Invitrogen, Carlsbad CA), followed by treatment with DNA-Free reagent (Ambion, Austin TX) to remove con-
taminating DNA. Total RNA was then subjected to reverse transcription using an Omniscript RT kit with ran-
dom primers (Applied Biosystems, Foster CA). Quantitative real-time PCR analysis was carried out using an
ABI PRISM 7500 Sequence Detection System (Applied Biosystems Foster CA) with SYBR Green (Takara Bio,
Tokyo, Japan), and values were normalized with respect to β-actin. We designed primers which detect both Saa1
and Saa2 mRNA. The following primers were used: Saa1/2-F: 5′-AGTGGCAAAGACCCCAATTA-3′, Saa1/2-R:
5′-GGCAGTCCAGGAGGTCTGTA-3′; β-actin-F: 5′-GACAGGATGCAGAAGGAGATTACT-3′ and β-actin-R:
5′-TGATCCACATCTGCTGGAAGGT-3′.

Serum lipoprotein quantity, particle size, and distribution of apolipoproteins.  To determine

HDL particle size, serum (5 μl for C57BL/6 J mice and Apoa2−/− mice) samples were pre-stained for lipids with
Sudan Black B and electrophoresed on a non-denaturing PAGE gel with a 5 to 15% linear polyacrylamide gra-
dient. Electrophoresis was carried out at 25 mA for 2 h32,33. The distribution of ApoA-I, ApoA-II, ApoE, and
SAA protein among the various lipoprotein particles was determined by Western blot analysis of 1 μl serum
separated by non-denaturing PAGE. Antibodies used included: anti-mouse AA (1:3000)62, rabbit anti-mouse
ApoA-I (1:3000)65,66, rabbit anti-mouse ApoA-II (1:200) (sc-366255, Santa Cruz Biotechnology, Dallas, TX) and
goat anti-apoE antiserum (1:200) (sc-6384, Santa Cruz Biotechnology). Subsequently, membranes were incubated
with anti-rabbit IgG solution (1:3000) or donkey anti-goat IgG-peroxidase conjugated solution (1:3000) (sc-2020,
Santa Cruz Biotechnology). To further determine the cholesterol profiles in serum lipoproteins, pooled sera from
five C57BL/6 J mice and six Apoa2−/− mice treated with AgNO3 and AA fibrils were analyzed by dual-detection
high-performance liquid chromatography (HPLC) (Liposearch System, Skylight Biotech, Inc., Akita, Japan)67.
In addition, isolated HPLC fractions with different particle diameters (ranging from 7.6 nm to >80 nm) from
30 µl pooled serum were concentrated. SAA protein in each fraction was detected by Western blot analysis after
separation by 16.5% SDS-PAGE.

Statistical analysis.  The SPSS 16.0 software package (Abacus Concepts, Berkley, CA) was used to analyze
the data. All data are presented as mean ± SEM. Significant differences in real time PCR and AA IHC positive
areas were examined by unpaired Student’s t-test or the Tukey–Kramer method for multiple testing. Because
the AI and damage score are nonlinear indexes, significant differences were examined using the nonparametric
Mann-Whitney U-test. A two-tailed P value of <0.05 was considered to be statistically significant.

References
1. Merlini, G. & Bellotti, V. Molecular mechanisms of amyloidosis. N Engl J Med 349, 583–596 (2003).
2. Westermark, P. et al. Amyloid fibril protein nomenclature–2002. Amyloid 9, 197–200 (2002).
3. Sipe, J. D. et al. Amyloid fibril proteins and amyloidosis: chemical identification and clinical classification International Society of
Amyloidosis 2016 Nomenclature Guidelines. Amyloid 23, 209–213 (2016).
4. Lachmann, H. J. et al. Natural history and outcome in systemic AA amyloidosis. N Engl J Med. 356, 2361–71 (2007).
5. Jakob, W. Spontaneous amyloidosis of mammals. Vet Pathol 8, 292–306 (1971).
6. Landman, W. J. et al. Avian amyloidosis. Avian Pathol 27, 437–449 (1998).
7. Marhaug, G. & Dowton, S. B. Serum amyloid A: an acute phase apolipoprotein and precursor of AA amyloid. Baillieres Clin.
Rheumatol 8, 553–573 (1994).
8. Kisilevsky, R. & Manley, P. N. Acute-phase serum amyloid A: perspectives on its physiological and pathological roles. Amyloid 19,
5–14 (2012).
9. Hazenberg, B. P. & van Rijswijk, M. H. Where has secondary amyloid gone? Ann Rheum Dis. 59, 577–579 (2000).
10. Nakamura, T. Amyloid A amyloidosis secondary to rheumatoid arthritis: uncommon yet important complication. Curr Rheumatol
Rev 3, 231–241 (2007).
11. Benditt, E. P. & Eriksen, N. Amyloid protein SAA is associated with high density lipoprotein from human serum. Proc. Natl. Acad.
Sci. USA 74, 4025–4028 (1977).
12. Skogen, B. et al. A high molecular weight serum protein is the carrier for amyloid-related protein SAA. Scand. J. Immunol 6,
1363–1368 (1977).
13. Hoffman, J. S. et al. Murine tissue amyloid protein AA. NH2-terminal sequence identity with only one of two serum amyloid protein
(ApoSAA) gene products. J Exp Med 159, 641–646 (1984).
14. Baranova, I. N. et al. CD36 is a novel serum amyloid A (SAA) receptor mediating SAA binding and SAA-induced signaling in
human and rodent cells. J Biol Chem. 285, 8492–8506 (2010).
15. Cheng, N. et al. Cutting edge: TLR2 is a functional receptor for acute-phase serum amyloid A. J Immunol 181, 22–26 (2008).
16. Hoffman, J. S. & Benditt, E. P. Changes in high density lipoprotein content following endotoxin administration in the mouse:
formation of serum amyloid protein-rich subfractions. J Biol Chem 257, 10510–10517 (1982).

SCIeNtIfIC REPOrTS | (2018) 8:5620 | DOI:10.1038/s41598-018-23755-y 9

www.nature.com/scientificreports/

17. Ganowiak., K. et al. Fibrils from synthetic amyloid-related peptides enhance development of experimental AA-amyloidosis in mice.
Biochem Biophys Res Commun 199, 306–12 (1994).
18. Murakami, T. et al. Experimental induction and oral transmission of avian AA amyloidosis in vaccinated white hens. Amyloid 20,
80–5 (2013).
19. Horiuchi, N. et al. Experimental induction of amyloidosis by bovine amyloid fibrils in sore hock rabbits. Amyloid 15, 84–88 (2008).
20. Escola-Gil, J. C. et al. Antiatherogenic role of high-density lipoproteins: insights from genetically engineered-mice. Front Biosci 11,
1328–1348 (2006).
21. David, O. et al. Apolipoprotein mimetic peptides: Mechanisms of action as anti-atherogenic agents. Pharmacol Ther 130, 83–91
(2011).
22. Blanco-Vaca., F. et al. Role of apoA-II in lipid metabolism and atherosclerosis: advances in the study of an enigmatic protein. J. Lipid
Res 42, 1727–1739 (2001).
23. Catherine, C. H. et al. In vivo interactions of apoA-II, apoA-I, and hepatic lipase contributing to HDL structure and antiatherogenic
functions. J Lipid Res 42, 563–570 (2001).
24. Shaobin Zhong. et al. Human ApoA-II Inhibits the hydrolysis of HDL triglyceride and the decrease of HDL size Induced by
hypertriglyceridemia and cholesteryl ester transfer Protein in transgenic mice. J Clin Invest 94, 2457–2467 (1994).
25. Hedrick, C. C. et al. Influence of mouse apolipoprotein A-II on plasma lipoproteins in transgenic mice. J Biol Chem 268,
20676–20682 (1993).
26. Yonezu, T. et al. High homology is present in the primary structures between murine senile amyloid protein (ASSAM) and human
apolipoprotein A-II. FEBS Lett 203, 149–52 (1986).
27. Higuchi., K. et al. Polymorphism of apolipoprotein A-II (apoA-II) among inbred strains of mice. Relationship between the molecular
type of apoA-II and mouse senile amyloidosis. Biochem J 279, 427–33 (1991).
28. Yan, J. et al. Cross-seeding and crosscompetition in mouse apolipoprotein A-II amyloid fibrils and protein A amyloid fibrils. Am J
Pathol 171, 172–80 (2007).
29. Wang, Y. et al. ApoA-I deficiency in mice is associated with redistribution of apoA-II and aggravated AApoAII amyloidosis. J Lipid
Res 52, 1461–1470 (2011).
30. Eriksen, N. & Benditt, E. P. Isolation and characterization of the amyloid-related apoprotein (SAA) from human high density
lipoprotein. Proc Natl Acad Sci USA 77, 6860–6864 (1980).
31. Ashby, D. et al. Lack of effect of serum amyloid A (SAA) on the ability of high-density lipoproteins to inhibit endothelial cell
adhesion molecule expression. Atherosclerosis. 154, 113–121 (2001).
32. Fengxia Ge. et al. Amyloidosis in transgenic mice expressing murine amyloidogenic apolipoprotein A-II (Apoa2c). Laboratory
Investigation 87, 633–643 (2007).
33. Umezawa, M. et al. Dietary fat modulation of apoA-II metabolism and prevention of senile amyloidosis in the senescence-
accelerated mouse. J. Lipid Res 44, 762–769 (2003).
34. Castellani, L. W. et al. Overexpression of apolipoprotein AII in transgenic mice converts high density lipoproteins to
proinflammatory particles. J Clin Invest 100, 464–474 (1997).
35. Thompson, P. A. et al. Apolipoprotein A-II augments monocyte responses to LPS by suppressing the inhibitory activity of LPS-
binding protein. Innate Immun 14, 365–374 (2008).
36. Koohdani, F. et al. APOA2 -265T/C polymorphism is associated with increased inflammatory responses in patients with type 2
diabetes mellitus. Diabetes Metab J 40, 222–229 (2016).
37. Westermark, G. T. et al. AA amyloidosis: pathogenesis and targeted therapy. Annu Rev Pathol 10, 321–344 (2015).
38. Lundmark, K. et al. Depletion of spleen macrophages delays AA amyloid development: a study performed in the rapid mouse model
of AA amyloidosis. PLoS One 8, e79104 (2013).
39. Sponarova, J. et al. Efficient amyloid A clearance in the absence of immunoglobulins and complement factors. Am J Pathol 182,
1297–307 (2013).
40. Simons, J. P. et al. Pathogenetic mechanisms of amyloid A amyloidosis. Proc Natl Acad Sci USA 110, 16115–16120 (2013).
41. Solomon, A. et al. Transgenic mouse model of AA amyloidosis. Am J Pathol 154, 1267–1272 (1999).
42. Lin Li. et al. Caloric restriction reduces the systemic progression of mouse AApoAII amyloidosis. PLoS One 12, e0172402 (2017).
43. Khovidhunkit, W. et al. Effects of infection and inflammation on lipid and lipoprotein metabolism: mechanisms and consequences
to the host. J. Lipid Res 45, 1169–1196 (2004).
44. Trieb, M. et al. Liver disease alters high-density lipoprotein composition, metabolism and function. Biochim Biophys Acta 1861,
630–638 (2016).
45. Emancipator, K. et al. In vitro inactivation of bacterial endotoxin by human lipoproteins and apolipoproteins. Infect Immun 60,
596–601 (1992).
46. Flegel, W. A. et al. Prevention of endotoxin-induced monokine release by human low- and high-density lipoproteins and by
apolipoprotein A-I. Infect Immun 61, 5140–5146 (1993).
47. Ma, J. et al. Role of apolipoprotein A-I in protecting against endotoxin toxicity. Acta Biochim Biophys Sin 36, 419–424 (2004).
48. Sammalkorpi, K. et al. Changes in serum lipoprotein pattern induced by acute infections. Metabolism 37, 859–865 (1988).
49. Grunfeld, C. et al. Lipids, lipoproteins, triglyceride clearance, and cytokines in human immunodeficiency virus infection and the
acquired immunodeficiency syndrome. J. Clin. Endocrinol. Metab 74, 1045–1052 (1992).
50. Feingold, K. R. et al. Effect of endotoxin on cholesterol biosynthesis and distribution in serum lipoproteins in Syrian hamsters. J.
Lipid Res 34, 2147–2158 (1993).
51. Cabana, V. G. et al. Effects of the acute phase response on the concentration and density distributionnof plasma lipids and
apolipoproteins. J. Lipid Res 30, 39–49 (1989).
52. Clifton, P. M. et al. Effects of serum amyloid A protein (SAA) on composition, size, and density of high density lipoproteins in
subjects with myocardial infarction. J. Lipid Res 26, 1389–1398 (1985).
53. Hosoai, H. et al. Expression of serum amyloid A protein in the absence of the acute phase response does not reduce HDL cholesterol
or apoA-I levels in human apoA-I transgenic mice. J. Lipid Res 40, 648–653 (1999).
54. Weng, W. & Breslow, J. L. Dramatically decreased high density lipoprotein cholesterol, increased remnant clearance, and insulin
hypersensitivity in apolipoprotein A-II knockout mice suggest a complex role for apolipoprotein A-II in atherosclerosis
susceptibility. Proc. Natl. Acad. Sci. USA 93, 14788–14794 (1996).
55. Doolittle, M. H. et al. A polymorphism affecting apolipoprotein A-II transla- tional efficiency determines high density lipoprotein
size and composition. J. Biol. Chem 265, 16380–16388 (1990).
56. Scanu, A. M. & Edelstein, C. HDL: bridging past and present with a look at the future. FASEB J 22, 4044–4054 (2008).
57. Maïga, S. F. et al. Apolipoprotein A-II is a key regulatory factor of HDL metabolism as appears from studies with transgenic animals
and clinical outcomes. Biochimie 96, 56–66 (2014).
58. Gao, X. et al. Role of apolipoprotein A-II in the structure and remodeling of human high-density lipoprotein (HDL): protein
conformational ensemble on HDL. Biochemistry 51, 4633–4641 (2012).
59. Frame, N. M. & Gursky, O. Structure of serum amyloid A suggests a mechanism for selective lipoprotein binding and functions: SAA
as a hub in macromolecular interaction networks. FEBS Lett 590, 866–879 (2016).
60. Koike, T. et al. Expression of human apoAII in transgenic rabbits leads to dyslipidemia: a new model for combined hyperlipidemia.
Arterioscler. Thromb. Vasc. Biol 29, 2047–2053 (2009).

SCIeNtIfIC REPOrTS | (2018) 8:5620 | DOI:10.1038/s41598-018-23755-y 10

www.nature.com/scientificreports/

61. Puchtler, H. et al. On the binding of Congo red by amyloid. J. Histochem Cytochem 10, 355–364 (1962).
62. Higuchi, K. et al. Systemic senile amyloid in senescence-accelerated mice: a unique fibril protein demonstrated in tissues from
various organs by the unlabeled immunoperoxidase method. Lab Invest 48, 231–240 (1983).
63. Xing, Y. et al. Induction of protein conformational change in mouse senile amyloidosis. J Biol Chem 277, 33164–33169 (2002).
64. Sawashita, J. et al. C-terminal sequence of amyloid- resistant type F apolipoprotein A-II inhibits amyloid fibril formation of
apolipoprotein A-II in mice. Proc Natl Acad Sci USA 112, 836–845 (2015).
65. Chiba, T. et al. Mouse senile amyloid deposition is suppressed by adenovirus-mediated overexpression of amyloid-resistant
apolipoprotein A-II. Am J Pathol 155, 1319–1326 (1999).
66. Higuchi, K. et al. Age-related changes of serum apoprotein SASSAM, apoprotein A-I and low-density lipoprotein levels in
senescence accelerated mouse (SAM). Mech Ageing Dev 26, 311–26 (1984).
67. Okazaki, M. et al. Identification of unique lipoprotein subclasses for visceral obesity by component analysis of cholesterol profile in
high-performance liquid chromatography. Arterioscler. Thromb. Vasc. Biol 25, 578–584 (2005).

Acknowledgements
This work was supported in part by Grants-in-Aid for Scientific Research (B) 26293084, 17H04063 and
Challenging Exploratory Research 26670152 from the Ministry of Education, Culture, Sports, Science and
Technology, Japan. The authors thank Drs Kiyoshi Matsumoto and Takahiro Yoshizawa, Ms. Kayo Suzuki
(Research Center for Supports to Advanced Science, Shinshu University) for animal care and technical assistance
for making tissue sections.

Author Contributions
M.Y., J.S. and K.H. conceived and designed the experiments, M.Y., Y.L., J.D., L.L., X.D. and X.Z. performed
experiments and was responsible for data acquisition and analysis, J.D., H.M. and X.D. analyzed the data, M.M.
interpreted the data and provided the experimental methods, M.Y., S.J. and K.H. wrote the manuscript, and all
authors reviewed the manuscript.

Additional Information
Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-018-23755-y.
Competing Interests: The authors declare no competing interests.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

© The Author(s) 2018

SCIeNtIfIC REPOrTS | (2018) 8:5620 | DOI:10.1038/s41598-018-23755-y 11